Factorial structure and measurement invariance of the Chinese version of the Oral Health Impact Profile-14 among clinical populations and non-clinical populations: an evidence for public oral investigations

Table 4 Measurement invariance model between different samples fitting indices and comparison

Model	χ² (df), p	CFI	TLI	SRMR	RMSEA	Model comparison	∆CFI
The clinical populations	129.044(56), p < 0.001	0.970	0.952	0.029	0.052
The non-clinical populations	102.076(56), p < 0.001	0.987	0.978	0.023	0.030
Model 1	229.067(112), p < 0.001	0.981	0.969	0.025	0.039
Model 2	242.511(119), p < 0.001	0.980	0.970	0.029	0.038	2 vs. 1	-0.001	-0.001
Model 3	259.125(126), p < 0.001	0.979	0.969	0.030	0.039	3 vs. 2	-0.001	0.001
Model 4	375.653(140), p < 0.001	0.962	0.951	0.038	0.049	4 vs. 3	-0.017	0.010

Model 1 = configural invariance; Model 2 = metric invariance; Model 3 = scalar invariance; Model 4 = error variance invariance; χ², chi-square; df, degree of freedom; CFI, comparative fit index; TLI, Tucker-Lewis index; SRMR, standardized root mean square residual; RMSEA, root-mean-square error of approximation; ∆, change in the parameter

ISSN: 1472-6831